Skip to main content
. 2023 Sep 2;30(12):2072–2082. doi: 10.1093/jamia/ocad168

Table 1.

Description of the parameters related to the dataset information.

Parameter Description
Country of data source
 United States 50 (61.7)
 China 13 (16.0)
 Others 15 (18.5)
 Not reported 3 (3.7)
Follow-up time, in years [0.8, 24], 8.9 (5.6), 10 [4–12], (70.4)
Sample size [398, 29 163 297], 585 981 [3 611 801], 32 221 [7845–105 805], (18.5)
Demographic descriptors
 Median age [29, 74.6], 55.6 (13.4), 56.1 [46.2–65.4], (23.5)
 Sex (female) [0, 100], 51 (20.6), 49.8 [43–61.4], (29.6)
 Socioeconomic information reported 6 (7.4)
Age limit definition 19 (23.5)
Main condition definition 27 (8.6)
Information included
 Demographics 39 (48.2)
 Diagnoses 60 (74.1)
 Laboratory results 38 (46.9)
 Prescribed/billed drugs 43 (53.1)
 Medical procedures 31 (38.3)
 Clinical measures 31 (38.3)
 Information included simultaneously (n) [1, 6], 3 (1.3), 3 [2–4], (98.8)
 Others (eg, free text, genes…) 17 (21)
Types of variables
 All quantitative 7 (8.6)
 All categorical 39 (48.1)
 Both 35 (43.2)

Categorical parameters are described as N (%), while quantitative parameters as [min, max], mean (SD), median [Q1-Q3], (% studies reported). Total N = 81. If more than one dataset was used to train the models, only the parameters of the biggest dataset were collected.